This proposal will enable the applicant to become an independent researcher in neuroimaging applied to aging and dementia. The cellular basis of functional imaging abnormalities is not clearly understood. Functional imaging offers a window on in vivo metabolism that can be correlated with pathology when tissue is available near the time of scanning. Postmortem studies have difficulty distinguishing dysfunctional neurons from dead neurons but offer fully accessible tissue for molecular analysis. The localization of postmortem tissue samples to premortem regions assessed by functional imaging can reveal the cellular basis of metabolic abnormalities. Sampling tissue relatively active on premortem scanning will avoid studying regions long destroyed by a pathologic process. The acute destructive process in Alzheimer's disease (AD) can be assessed by combining functional imaging and selective tissue analysis in patients close to death. Understanding the cellular basis of functional imaging changes in AD may assist the interpretation of imaging abnormalities in other diseases. Amyloid beta protein (ABeta) may contribute to the pathophysiology of AD-the mechanism may involve oxidative damage contributing to neuron and synapse loss; other mechanisms may also be contributory. [18F]-fluorodeoxyglucose positron emission tomography (FDG-PET) reflects energy metabolism at the synapse. Quantification of synaptophysin from postmortem tissue sections, obtained after scanning and near the time of death, can accurately reflect the density of synapses in a given PET voxel. This proposal details a plan for didactic study, acquiring laboratory skills, and supervised research to address the following hypothesis: Hypometabolism of FDG-PET in AD, when corrected for atrophy and synapse quantity, will show a significant correlation with ABeta and markers of oxidative damage: mitochondrial DNA deletions, lipid peroxidation, and decreased glucose-6-phosphate dehydrogenase. To test this hypothesis FDG-PET scans, obtained close to death in AD patients, will be coregistered with a 3D reconstruction of their ex vivo cryomacrotome image data. This will allow the pursuit of the following specific aims: (1) Correct the partial volume error of FDG-PET using high resolution cryomacrotome data and synaptophysin measures to arrive at corrected metabolic values. (2) Map the relationship between indicators of oxidative damage with the corrected premortem functional data. (3) Integrate the individual patient maps into a population map.